Aglet

Investigate Missing and Zero Report States

A missing-versus-zero investigation needs one period with no row, one explicit zero, one positive value, and one unavailable calculation if the report supports it. Rebuild the measure independently, then compare visible and exported surfaces. The brief should name the first state change and keep source absence distinct from formatting.

Build a useful investigation brief

  1. Build a state fixture

    Create rows or period buckets for no observation, explicit zero, positive value, null measure, and unsupported calculation as applicable. Record expected table cell, chart point, card text, export value, and filter scope for every state.

  2. Follow each state

    Capture population after filtering and grouping, aggregate output, missing-value handling, chart data, formatting, and serialized export. Compare source membership with visible output. Mark the first stage that invents zero, suppresses a point, or labels an unavailable result as blank.

  3. Test one display rule

    Change only the missing-value policy, such as preserve blank, show no data, or substitute zero under a contract. Rerun the fixture across a card, table, chart, and export. Treat a semantic disagreement as a policy hypothesis until the metric owner confirms the intended reading.

What to carry forward

The investigation is ready when the brief names the state fixture, first semantic collapse, affected surfaces, strongest evidence, and competing display rule. Recommend a scoped measure or presentation correction. Keep the case open when source absence and explicit zero are not observable separately.

Technical background: Metabase documentation.

Keep the decision with the work.

Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.

Create an account See the product workflow